Up to 75% of breast tumors are estrogen receptor (ER) positive and nearly 50% of these will recur following endocrine therapy. Thus, new therapeutic targets and strategies are urgently needed for a large population of women. Our work demonstrates that ER and NF?B interact cooperatively to synergistically up-regulate a gene signature that (i) is correlated with the more aggressive, Luminal B phenotype of ER+ breast cancer and (ii) predicts an increased risk of recurrence for women given endocrine therapy. Moreover, our preliminary studies suggest that an expansion of the stem-like or tumor-initiating cell (BCSC) population is a response to increased ER/NF?B crosstalk resulting in TAM-resistance and subsequent tumor recurrence. We found that ER/NF?B crosstalk i) increases expression of multiple genes associated with BCSCs, ii) promotes expansion of a cell population expressing BCSC markers, and iii) enhances mammosphere formation, a functional readout of BCSC activity. Importantly, we identified PHLDA1 (Pleckstrin homology-like domain A1) as a key ER/NF?B target gene that is not only necessary for these phenotypes but is also up-regulated in both isolated BCSCs and ER+ tumors in patients that respond poorly to TAM. Based on these findings, we hypothesize that the coordinated activity of ER and NF?B promotes expansion of the BCSC population in ER+ breast cancers and that this drives progression to a TAM-resistant and recurrent phenotype. Moreover, we propose ER/NF?B target genes, such as PHLDA1, are critical mediators of these ER/NF?B driven phenotypes. In Aim 1, we will test the hypothesis that ER/NF?B cooperatively enables expansion of the BCSC population in ER+ breast cancer cell lines and patient-derived tumors. Whether this expansion is through BCSC self-renewal and is BCSC-intrinsic will be investigated. In Aim 2, we will investigate how activation of the ER/NF?B axis in breast cancer cells and tumors promotes the development of SERM resistance in vitro and tumor recurrence in vivo. Whether depletion or targeting of BCSCs can prevent TAM resistance and tumor recurrence will be investigated. And in Aim 3, we will identify critical ER/NFkB crosstalk genes, such as PHLDA1, that are necessary for the maintenance and/or expansion of the BCSC population of ER+ breast cancers. Whether these factors also contribute to the development of SERM resistance and tumor recurrence will be determined. From the proposed studies, we expect to provide the rationale for developing new and effective therapeutic strategies to target ER-NF?B crosstalk and BCSC activity in ER+ tumors. Our long-term goal is to use these strategies to prevent tumor relapse and progression in women with ER+ breast cancers on endocrine therapy.